Website update

Text view on website, Mobilizer, Text Bookmarklet:

• Added font controls. You can now change the font, font size, line height, and margin (margin not available on iPhone due to impracticality). Settings are saved in a cookie so they stick even when you aren’t logged in, like when using the Mobilizer in other applications.

API:

• By default, all URLs added will be crawled synchronously to find the final URL after redirects have been followed. Other normalization may also be applied. This should prevent URLs from being saved with URL-shortener hosts (like bit.ly) and RSS redirects (like feedproxy.google.com). Duplicate-checking applies to the final URL, not the originally submitted URL, when they differ.
• The auto-title functionality, in which Instapaper crawls the target URL to detect its title, is now used automatically if you do not specify a title parameter, or if the title parameter is an empty string. Previously, it was off unless specified. As a result, auto-title=1 is now redundant and deprecated, and it has been removed from the documentation.
• Added HTTP Basic Auth as an alternative to the username and password POST parameters. Please use Basic Auth whenever possible.
• The Content-Location header in the output previously used the instapaper.com/go/(ID) URLs. It now uses the item’s target URL after redirects and normalization.

Page compatibility:

• Added to “save full contents” login-requiring-site list: livejournal.com, stratfor.com, soundonsound.com. Raised saved-HTML size limit to accommodate more sites on the list.
• Fixed a bug that prevented americansuburbx.com and a few other sites from working in the text parser.
• Fixed a filter that occasionally caused a stray “<” symbol to remain at the end of text versions of certain pages.
• Fixed a bug in which IDN conversion was incorrectly applied to WordPress slugs with extended characters.

Some overdue site cleanup:

• Removed “Automatically move articles to the Archive when I read them” option and behavior. Its convenience was outweighed by its poor usability and resulting confusion among users. The /go URLs will start to be removed where they’re no longer necessary to improve link-clickthrough speed and prevent user errors.
• Removed option to select the old text parser. Only the new text parser is used from this point forward.

Please let me know if you see anything that isn’t working as it should be. Thanks.